|
|
|
AUTHOR(S): Sallama Resen
|
|
TITLE Smartphone Video Motion Deblur Modeled Based on Estimation Blur Parameter |
 PDF |
|
ABSTRACT many research approaches focus on image processing on smartphone platforms, visual object, augmented reality, object detection, tracking and recognition. sensor sizes are the main differences between camera smartphone and digital camera where sensor size is smaller in smartphone camera. image quality direct proportion with sensor size wherever the bigger get high quality. Most off smartphones featuring multiple cameras, multiple sensors in one device, but still none of them having large as digital camera. blur occur when there is relative motion between the camera and the object scene while capturing the image. blur, typically occur in low-light scenarios due to requiring exposures. lens blur doesn't change significantly all over the image while object motion is highly directional and changes abruptly depending on the objects. Paper present method deal with the main challenges occur in smartphone video platform. proposed method eliminates small motion blur in three phases. Blur estimation achieved by prior information on distribution image gradient. Orientation Gaussian Filter fit the prior information to find the regression coefficients. Multi order combine different estimate GOF parameters to generate removal blur filter. Estimation parameters are fix and set blur on the image to produce image without boosting the noise and unwanted artifacts. Proposed model generated images that have more details instead of directly minimize which is solve optimization problem by minimize loss function. Suggested method applies on outdoor, indoor video acquired by modern smartphone. Experiment result display is accurately for the full regression motion blur model. suggested model exam on video dataset conditions 23:75 sec, 229,44 MP. measurement evaluation established on time consumer, SSIM and PSNR. experimental results show image artifacts phase less consuming computational time. proposed model has minimized cost function and form image quality greater. |
|
KEYWORDS Smartphone Platform, Motion Blur, Orientation Gaussian, Blur Filter |
|
|
|
Cite this paper Sallama Resen. (2024) Smartphone Video Motion Deblur Modeled Based on Estimation Blur Parameter. International Journal of Communications, 9, 5-10 |
|
|
